Xin Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Xin Liu has a total of 96 co-authors. Here are the top co-authors by number of shared publications:
- Wenhua Chen
Tsinghua University
26 shared publications - Zhenghe Feng
Department of Electronic Engineering, Tsinghua University, Beijing, China
17 shared publications - Ting Feng
Xidian University
16 shared publications - Xiaohua Ma
Xidian University
16 shared publications - Yang Lu
Xidian University
16 shared publications - Chupeng Yi
Xidian University
14 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 74 (11) , 9681-9693
A 20--28-GHz Balanced Front End Using 3-dB Quadrature-Coupler Impedance Transformation in 65-nm CMOS
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 74 (11) , 9681-9693
A 20--28-GHz Balanced Front End Using 3-dB Quadrature-Coupler Impedance Transformation in 65-nm CMOS
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 74 (11) , 9669-9680
A W-Band Power Amplifier With Vector-Sum Equalizer Technique Achieving 43% Fractional Bandwidth
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 74 (11) , 9669-9680
A W-Band Power Amplifier With Vector-Sum Equalizer Technique Achieving 43% Fractional Bandwidth
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 45 (11) , 5521-5529
A High-Accuracy, Hybrid Scalable Model for MIM Capacitors Over Wide Temperature for MMIC Applications
Conventional Process Design Kits for Monolithic Microwave Integrated Circuits (MMICs) often lack capacitor models that are robust across wide temperature and frequency ranges, posing a significant challenge to the design reliability and first-pass success. To address this, a novel π-type equivalent circuit with grounding inductors is developed, yielding S-parameter fitting errors under 2% while also explaining the device’s thermal resonant drift. Building on this, a hybrid scalable model synergizing a physics-guided core with neural network compensation is proposed. This methodology demonstrates exceptional accuracy, achieving parameter-fitting R-squared values above 0.999. More importantly, the final model’s S-parameter predictions maintain an R-squared greater than 0.98 against measured data across the entire temperature and geometry space. This work provides a highly accurate, scalable MIM capacitor model essential for robust circuit design under harsh operating conditions and offers a generalizable hybrid methodology applicable to other passive components, enhancing the predictive power of modern PDKs.
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 45 (11) , 5521-5529
A High-Accuracy, Hybrid Scalable Model for MIM Capacitors Over Wide Temperature for MMIC Applications
Conventional Process Design Kits for Monolithic Microwave Integrated Circuits (MMICs) often lack capacitor models that are robust across wide temperature and frequency ranges, posing a significant challenge to the design reliability and first-pass success. To address this, a novel π-type equivalent circuit with grounding inductors is developed, yielding S-parameter fitting errors under 2% while also explaining the device’s thermal resonant drift. Building on this, a hybrid scalable model synergizing a physics-guided core with neural network compensation is proposed. This methodology demonstrates exceptional accuracy, achieving parameter-fitting R-squared values above 0.999. More importantly, the final model’s S-parameter predictions maintain an R-squared greater than 0.98 against measured data across the entire temperature and geometry space. This work provides a highly accurate, scalable MIM capacitor model essential for robust circuit design under harsh operating conditions and offers a generalizable hybrid methodology applicable to other passive components, enhancing the predictive power of modern PDKs.
- 18 August 2026
- Published by Wiley in Microwave and Optical Technology Letters
- Vol. 68 (8) , e70750
A Systematic Design of an Ultra‐Compact and Low‐Loss Broadband Filtering Power Divider Using a Novel Three‐Conductor Broadside Coupled Line in Substrate Integrated Suspended Line
Designing broadband filtering power dividers with simultaneous wide bandwidth, low insertion loss, and compact size remains a significant challenge. This paper presents a novel solution using a three‐conductor broadside coupled line (TCBCL) structure, implemented on a low‐loss multilayer substrate integrated suspended line (SISL) platform. The proposed design features an air‐gap quasi‐metallic configuration between conductors to minimize dielectric loss. Input power is coupled directly from the center conductor to two side conductors, achieving equal power division ratio with an inherent broadband bandpass response. The output ports isolation is enhanced by vertically connecting two surface mounted resistors between the top and bottom conductors. For validation, a prototype was fabricated and measured, exhibiting an ultra‐compact core circuit size of only 4.11 mm × 1.08 mm × 3.25 mm. Measured results show an insertion loss below 0.57 dB across 7.7–18.1 GHz, corresponding to a fractional bandwidth (FBW) of 80.62%. This work demonstrates an effective and miniaturized design approach for high‐performance, highly integrated RF front‐ends and passive devices.
- 18 August 2026
- Published by Wiley in Microwave and Optical Technology Letters
- Vol. 68 (8) , e70750
A Systematic Design of an Ultra‐Compact and Low‐Loss Broadband Filtering Power Divider Using a Novel Three‐Conductor Broadside Coupled Line in Substrate Integrated Suspended Line
Designing broadband filtering power dividers with simultaneous wide bandwidth, low insertion loss, and compact size remains a significant challenge. This paper presents a novel solution using a three‐conductor broadside coupled line (TCBCL) structure, implemented on a low‐loss multilayer substrate integrated suspended line (SISL) platform. The proposed design features an air‐gap quasi‐metallic configuration between conductors to minimize dielectric loss. Input power is coupled directly from the center conductor to two side conductors, achieving equal power division ratio with an inherent broadband bandpass response. The output ports isolation is enhanced by vertically connecting two surface mounted resistors between the top and bottom conductors. For validation, a prototype was fabricated and measured, exhibiting an ultra‐compact core circuit size of only 4.11 mm × 1.08 mm × 3.25 mm. Measured results show an insertion loss below 0.57 dB across 7.7–18.1 GHz, corresponding to a fractional bandwidth (FBW) of 80.62%. This work demonstrates an effective and miniaturized design approach for high‐performance, highly integrated RF front‐ends and passive devices.
- 15 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. PP (00189480) , 1-18
Crosstalk-Mapping Digital Predistortion for Fully Connected Hybrid Beamforming MIMO Transmitters Using Piecewise Approximate Vector Synthesis
A low-complexity digital predistortion (DPD) technique is proposed to address the linearization issue in the fully connected hybrid beamforming (FC-HBF) multiple input multiple output (MIMO) transmitters. The proposed DPD model is developed based on a piecewise approximate vector synthesis (PAVS) approach, with third-order polynomials integrated to enhance the linearization accuracy. A comprehensive analysis of the radiation characteristics of the crosstalk components is conducted, enabling the model to activate relevant crosstalk mapping (CM)-intermodulation distortion (IMD) terms selectively. In this way, the proposed DPD technique can substantially reduce the computational complexity while preserving excellent linearization performance. The proposed DPD is evaluated through simulations and measurements against the state-of-the-art DPDs. Simulations on a $16\times 16$ FC-HBF array show that the proposed DPD achieves superior linearization performance while reducing complexity by about 90% for four-stream configuration. Proof-of-concept tests on a linear array to verify its effectiveness, indicating that the proposed DPD can achieve competitive linearization performance and less complexity in comparison to the other models. These results confirm the strong potential of the proposed DPD for effectively linearizing FC-HBF arrays.
- 15 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. PP (00189480) , 1-18
Crosstalk-Mapping Digital Predistortion for Fully Connected Hybrid Beamforming MIMO Transmitters Using Piecewise Approximate Vector Synthesis
A low-complexity digital predistortion (DPD) technique is proposed to address the linearization issue in the fully connected hybrid beamforming (FC-HBF) multiple input multiple output (MIMO) transmitters. The proposed DPD model is developed based on a piecewise approximate vector synthesis (PAVS) approach, with third-order polynomials integrated to enhance the linearization accuracy. A comprehensive analysis of the radiation characteristics of the crosstalk components is conducted, enabling the model to activate relevant crosstalk mapping (CM)-intermodulation distortion (IMD) terms selectively. In this way, the proposed DPD technique can substantially reduce the computational complexity while preserving excellent linearization performance. The proposed DPD is evaluated through simulations and measurements against the state-of-the-art DPDs. Simulations on a $16\times 16$ FC-HBF array show that the proposed DPD achieves superior linearization performance while reducing complexity by about 90% for four-stream configuration. Proof-of-concept tests on a linear array to verify its effectiveness, indicating that the proposed DPD can achieve competitive linearization performance and less complexity in comparison to the other models. These results confirm the strong potential of the proposed DPD for effectively linearizing FC-HBF arrays.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Microwave and Wireless Technology Letters
- Vol. 36 (6) , 898-901
Residual Structure-Based Multimodel Neural Network With Physically Inspired Core for Digital Predistortion in 6G Intelligent Radio
Digital predistortion (DPD) is the core technology for mitigating nonlinear distortion of power amplifiers (PAs) in the radio frequency (RF) front-end, especially in 6G integrated sensing and communication systems, where rapidly varying signal parameters demand real-time adaptive correction. To meet this challenge, this article proposes a dynamic DPD neural network (NN) model based on a multimodel residual structure NN (MRS-NN). Inspired by the PA’s bandpass equivalent circuits model, a low-complexity bandpass feedback recurrent NN (BPF-RNN) is developed as the core nonlinear fitting model in MRS-NN. Meanwhile, a multilayer fully connected network is integrated as the feature-extraction module to capture the dynamic characteristics of PA. Experimental results obtained over a range of operating scenarios—covering 64 trained and 16 untrained states—demonstrate that the proposed MRS-NN outperforms existing dynamic DPD models, delivering superior linearization performance while significantly reducing parameter count and computational complexity.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Microwave and Wireless Technology Letters
- Vol. 36 (6) , 898-901
Residual Structure-Based Multimodel Neural Network With Physically Inspired Core for Digital Predistortion in 6G Intelligent Radio
Digital predistortion (DPD) is the core technology for mitigating nonlinear distortion of power amplifiers (PAs) in the radio frequency (RF) front-end, especially in 6G integrated sensing and communication systems, where rapidly varying signal parameters demand real-time adaptive correction. To meet this challenge, this article proposes a dynamic DPD neural network (NN) model based on a multimodel residual structure NN (MRS-NN). Inspired by the PA’s bandpass equivalent circuits model, a low-complexity bandpass feedback recurrent NN (BPF-RNN) is developed as the core nonlinear fitting model in MRS-NN. Meanwhile, a multilayer fully connected network is integrated as the feature-extraction module to capture the dynamic characteristics of PA. Experimental results obtained over a range of operating scenarios—covering 64 trained and 16 untrained states—demonstrate that the proposed MRS-NN outperforms existing dynamic DPD models, delivering superior linearization performance while significantly reducing parameter count and computational complexity.